A Comprehensive Survey on AI-based Pest Detection in Stored Food Grains

Stored food commodities such as rice, broad beans, groundnut, and maize are vulnerable to pest infestations that cause significant post-harvest losses and threaten food security. Conventional inspection and monitoring techniques are often manual, time-intensive, and prone to human error. The emergence of Artificial Intelligence (AI) has enabled advanced solutions for pest detection and quality assessment in stored food products. Through machine learning (ML) and deep learning (DL) algorithms such as YOLO, Faster R-CNN, and Transformer-based architectures, intelligent systems can now accurately identify and classify pest species in varied storage conditions. These AI-driven models enhance early pest detection using image recognition, sensor data, and real-time analysis, thereby reducing chemical pesticide dependency and preserving product quality. In particular, lightweight and knowledgedistilled frameworks improve the feasibility of deploying detection systems in real-world storage environments for stored rice, broad beans, groundnut, and maize. Integration with IoT and federated learning enables distributed monitoring and privacy-preserving data handling across large-scale facilities. This survey comprehensively reviews AI-based methods for pest detection in stored agricultural products, analyzing their methodologies, datasets, performance, and challenges. The study also emphasizes the gaps in current research, such as dataset imbalance, model generalization, and deployment on low-power devices. Overall, this work underscores the pivotal role of AI in enhancing food safety, minimizing post-harvest losses, and supporting sustainable and intelligent storage management.